this post was submitted on 07 Jun 2023
134 points (93.5% liked)
Asklemmy
43963 readers
1700 users here now
A loosely moderated place to ask open-ended questions
Search asklemmy ๐
If your post meets the following criteria, it's welcome here!
- Open-ended question
- Not offensive: at this point, we do not have the bandwidth to moderate overtly political discussions. Assume best intent and be excellent to each other.
- Not regarding using or support for Lemmy: context, see the list of support communities and tools for finding communities below
- Not ad nauseam inducing: please make sure it is a question that would be new to most members
- An actual topic of discussion
Looking for support?
Looking for a community?
- Lemmyverse: community search
- sub.rehab: maps old subreddits to fediverse options, marks official as such
- !lemmy411@lemmy.ca: a community for finding communities
~Icon~ ~by~ ~@Double_A@discuss.tchncs.de~
founded 5 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Microservices and general "everything in the cloud" sentiment is stupid, it has ridiculous oerformance overheads and adds single points of failure that can easily prevent half the world from functioning.
I agree and I hate it. As much as possible I want to work with files locally.
I will disagree on microservices. I think it gets used for things that don't really require it like anything that gets hype, it gets overused. But I will give you example of exactly why and how it can be used correctly.
My company was early on the Kubernetes bandwagon. Before Amazon and Google provided hosted Kubernetes solutions, I had to write and bootstrap clusters automatically.
We had all our microservices deployed on k8s when covid hit. We went from about 4000k visitors/hour to 400,000k visitors an hour. This happened over a single week. k8s scaled out perfectly, our services scaled out perfectly and we had zero hiccups or downtime. We didn't have to do anything from an engineering perspective except increase our hard limits on the compute scale outs. During evening hours when load went down, the cluster would scale down.
If we were using a monolith application, we would have been really fucked. No way we could have scaled like that.